Sleep evaluation method, device and computer equipment

Abstract

The application discloses a sleep evaluation method and device and computer equipment, comprising: obtaining sleep monitoring data of a user to be evaluated; determining at least one continuity abnormal event based on the sleep monitoring data; calculating the event influence time of each continuity abnormal event based on the time attribute and characteristic attribute of each continuity abnormal event; determining the sleep continuity evaluation result of the user to be evaluated based on the event influence time of each continuity abnormal event. The computer equipment identifies different types of continuity abnormal events from the sleep monitoring data, and calculates the corresponding event influence time based on the time attribute and characteristic attribute of each continuity abnormal event, so that the influence of different abnormal conditions in the sleep process on the overall sleep is quantitatively integrated in the time dimension, and the sleep continuity evaluation result can more accurately reflect the actual degree of continuity damage in the sleep process, and the objectivity and rationality of sleep continuity evaluation are improved.

Description

Technical Field

[0001] This invention relates to the field of sleep data processing, and more particularly to a sleep assessment method, apparatus, and computer device. Background Technology

[0002] Sleep continuity is a crucial indicator of sleep quality, reflecting whether sleep stages remain relatively stable and whether they are frequently disturbed by awakenings or non-sleep states. Good sleep continuity typically helps restore bodily functions, while impaired sleep continuity can lead to daytime sleepiness, decreased attention, and various health risks. Therefore, objective and accurate assessment of sleep continuity is of great significance in the field of sleep health monitoring.

[0003] With the development of wearable devices and non-contact vital sign monitoring technologies, existing technologies are now able to acquire multi-dimensional sleep monitoring data, including sleep stage data and sleep event data. Among them, sleep stage data is typically used to characterize the sleep stage a user is in at different time periods, such as non-rapid eye movement (NREM) sleep, rapid eye movement (REM) sleep, and wakefulness; while sleep event data is used to mark specific events that occur during sleep, such as micro-awakening, wakefulness, or getting out of bed.

[0004] In existing sleep assessment methods, sleep continuity is usually assessed by counting the number of awakenings, the total duration of awakenings, or calculating the proportion of awake time in the entire night's sleep. However, these assessment methods tend to focus on counting the frequency or duration of events themselves and use uniform counting or weighting methods, making it difficult to accurately characterize the actual degree of sleep continuity impairment.

[0005] Therefore, how to make a more reasonable assessment of different types of continuity abnormalities during sleep based on existing sleep stage data and sleep event data remains a technical problem that urgently needs to be solved in the field of sleep continuity assessment. Summary of the Invention

[0006] Therefore, it is necessary to provide a sleep assessment method, device, and computer equipment to address the aforementioned technical problems and solve the difficulty of accurately characterizing sleep continuity in traditional sleep assessment methods.

[0007] A sleep assessment method, the method comprising:

[0008] Obtain sleep monitoring data from the user to be evaluated;

[0009] Based on the sleep monitoring data, at least one continuous abnormal event is identified. The continuous abnormal event is a micro-awakening event, a non-sleep event, or a duration abnormal event in which the duration of each sleep stage in the sleep monitoring data is abnormal.

[0010] Based on the time attribute and characteristic attribute of each of the continuous abnormal events, the event impact time of each of the continuous abnormal events is calculated;

[0011] The sleep continuity assessment result for the user to be evaluated is determined based on the event impact time of each of the aforementioned continuous abnormal events.

[0012] Optionally, the sleep monitoring data includes sleep event data, and the step of identifying at least one continuous abnormal event based on the sleep monitoring data includes:

[0013] The occurrence and end times of micro-awakening were determined from the sleep event data.

[0014] When the duration from the occurrence time point to the end time point is less than a preset micro-awakening duration, the sleep segment from the occurrence time point to the end time point is determined as the micro-awakening event.

[0015] Optionally, the sleep monitoring data includes sleep stage data, and the determination of at least one continuous abnormal event based on the sleep monitoring data includes:

[0016] Stage labels for each sleep stage are extracted from the sleep stage data;

[0017] If the stage label is not a preset sleep event label, then the corresponding sleep stage is the non-sleep event.

[0018] Optionally, determining at least one continuous abnormal event based on the sleep monitoring data further includes:

[0019] If the stage label is a preset sleep event label, then the stage duration corresponding to the sleep stage is determined;

[0020] Based on the duration of the aforementioned stage, a continuous duration deviation is judged in a preset reference duration dataset to obtain a judgment result;

[0021] When the determination result is that the duration of the stage is less than the corresponding reference duration in the preset reference duration dataset, the corresponding sleep stage is a duration abnormal event.

[0022] Optionally, before performing a continuous duration deviation judgment based on the stage duration in a preset reference duration dataset and obtaining the judgment result, the method further includes:

[0023] Obtain the user information of the user to be evaluated and historical sleep monitoring data that meet preset conditions;

[0024] Based on the historical sleep monitoring data, the reference duration calculation coefficients for each sleep stage were statistically obtained.

[0025] Based on the reference duration calculation coefficients for each sleep stage and the user information, the reference duration for each sleep stage is calculated.

[0026] The preset reference duration dataset is constructed based on the reference duration of each sleep stage.

[0027] Optionally, calculating the event impact time of each of the continuous abnormal events based on the time attributes and feature attributes of each event includes:

[0028] For the micro-awakening event, the time attribute of the micro-awakening event includes the sleep duration before the occurrence of the micro-awakening event and the duration of the micro-awakening event. The feature attribute includes the sleep cycle number where the micro-awakening event occurs, the degree of micro-awakening of the micro-awakening event, and the weight corresponding to the sleep stage before the occurrence of the micro-awakening event. The weight is set by the statistical probability of the micro-awakening event in each sleep stage.

[0029] The event impact time of the micro-awakening event is calculated based on the sleep duration before the micro-awakening event, the duration of the micro-awakening event, the sleep cycle number in which the micro-awakening event occurs, the degree of micro-awakening of the micro-awakening event, and the weight of the sleep stage before the micro-awakening event.

[0030] Optionally, calculating the event impact time of each of the continuous abnormal events based on the time attributes and feature attributes of each event includes:

[0031] For the non-sleep event, the time attribute includes the sleep duration before the non-sleep event occurred, and the feature attribute includes the sleep cycle number before the non-sleep event occurred.

[0032] The duration of sleep preceding the non-sleep event and the sleep cycle number preceding the non-sleep event are used to calculate the event impact time of the non-sleep event.

[0033] Optionally, calculating the event impact time of each of the continuous abnormal events based on the time attributes and feature attributes of each event includes:

[0034] For the duration abnormal event, the time attribute includes the actual duration of the sleep stage corresponding to the duration abnormal event and the reference duration of the sleep stage. The feature attribute includes the sleep cycle number of the sleep stage and the weight corresponding to the sleep stage. The weight is set according to the sleep depth of the user to be evaluated.

[0035] The event impact time of the duration abnormal event is calculated based on the actual duration of the sleep stage corresponding to the duration abnormal event, the reference duration of the sleep stage, the sleep cycle number of the sleep stage, and the weight of the sleep stage.

[0036] A sleep assessment device, the device comprising:

[0037] The first acquisition module is used to acquire sleep monitoring data of the user to be evaluated;

[0038] The first determining module is used to determine at least one continuous abnormal event based on the sleep monitoring data. The continuous abnormal event is a micro-awakening event, a non-sleep event, or a duration abnormal event of continuous duration of each sleep stage in the sleep monitoring data.

[0039] The first calculation module is used to calculate the event impact time of each of the continuous abnormal events based on the time attribute and feature attribute of each of the continuous abnormal events;

[0040] The second determining module is used to determine the sleep continuity assessment result of the user to be assessed based on the event impact time of each of the continuous abnormal events.

[0041] A computer device includes a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, wherein the processor implements the sleep assessment method described above when executing the computer-readable instructions.

[0042] To achieve the above-mentioned objectives, a readable storage medium is also provided, on which computer-readable instructions are stored, which, when executed by a processor, implement the sleep assessment method.

[0043] The aforementioned sleep assessment method, device, computer equipment, and storage medium acquire sleep monitoring data of the user to be assessed; based on the sleep monitoring data, identify at least one continuous abnormal event, which is a micro-awakening event, a non-sleep event, or a duration abnormality event indicating a continuous duration of abnormalities in each sleep stage in the sleep monitoring data; calculate the event impact time of each continuous abnormal event based on its temporal and characteristic attributes; and determine the sleep continuity assessment result of the user to be assessed based on the event impact time of each continuous abnormal event. The computer equipment, by performing the steps of identifying different types of continuous abnormal events from the sleep monitoring data and calculating the corresponding event impact time based on the temporal and characteristic attributes of each continuous abnormal event, quantifies and integrates the impact of different abnormalities on overall sleep in a time dimension. This avoids the shortcomings of rough assessments based solely on the number or duration of events, enabling the sleep continuity assessment results to more accurately reflect the actual degree of sleep continuity impairment, thus improving the objectivity and rationality of the sleep continuity assessment. Attached Figure Description

[0044] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0045] Figure 1 This is a flowchart illustrating a sleep assessment method according to an embodiment of the present invention;

[0046] Figure 2 This is one of the schematic diagrams of sleep staging results in one embodiment of the present invention;

[0047] Figure 3 This is a second schematic diagram of sleep staging results in one embodiment of the present invention;

[0048] Figure 4 This is a third schematic diagram of the sleep staging results in one embodiment of the present invention;

[0049] Figure 5 This is a schematic diagram of a sleep assessment device according to an embodiment of the present invention;

[0050] Figure 6 This is a schematic diagram of a computer device according to an embodiment of the present invention. Detailed Implementation

[0051] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0052] In one embodiment, such as Figure 1 As shown, a sleep assessment method is provided, including the following steps:

[0053] 101. Obtain sleep monitoring data of the user to be evaluated.

[0054] In this embodiment of the invention, the above-mentioned sleep assessment method can be deployed in a centralized sleep assessment platform, where the platform's computer equipment performs unified analysis and processing of the collected data, or it can be directly deployed in a smart bed, smart mattress, or other sleep monitoring device with local computing capabilities, where the device's processor executes a preset computer program to achieve localized sleep continuity assessment.

[0055] Users to be evaluated can be anyone who needs to analyze sleep quality or sleep continuity, such as people with long-term sleep disorders, ordinary users undergoing sleep health management, or individuals requiring sleep status assessment in medical, rehabilitation, or elderly care settings. This method is not limited by the user's age, gender, or usage scenario; as long as relevant sleep monitoring data can be obtained, the assessment can be completed.

[0056] Sleep monitoring data reflects changes in a user's state during sleep and can include sleep stage data and sleep event data. Sleep stage data describes the user's state at different time points, such as wakefulness, non-REM sleep, or REM sleep. Sleep event data records specific events that occur during sleep, such as micro-awakening, wakefulness, or getting out of bed. This data can be collected by wearable devices, bedside monitoring devices, or non-contact vital sign monitoring devices and stored in time-series format. Specifically, sleep monitoring data can be uploaded to a computer, where a processor executes computer programs to analyze and calculate the data, resulting in a sleep continuity assessment.

[0057] For example, during a night's sleep, the monitoring device continuously records changes in the user's sleep stages and generates corresponding event records when short-term awakenings are detected. These stage information and event records together constitute sleep monitoring data for subsequent continuous anomaly analysis.

[0058] 102. Based on sleep monitoring data, identify at least one continuous abnormal event.

[0059] In this embodiment of the invention, a continuous abnormal event is a micro-awakening event, a non-sleep event, or a duration abnormal event in which the duration of each sleep stage in the sleep monitoring data is abnormal.

[0060] Continuous abnormal events are used to describe situations that disrupt the normal sleep structure or reduce sleep stability during sleep. They mainly include three categories: micro-awakening events, non-sleep events, and duration abnormal events with continuous abnormal duration of sleep stages.

[0061] Micro-arousals refer to brief periods of wakefulness during sleep. These awakenings are short in duration, not reaching the level of full wakefulness or getting out of bed, but they can still disrupt the current sleep stage and subsequent sleep. Micro-arousals typically manifest as short, abrupt awakenings that occur suddenly during sleep, lasting from a few seconds to tens of seconds. Micro-arousals can be identified using sleep event data. When the duration corresponding to the start and end times of an awakening recorded in the monitoring data is less than a preset micro-arousal duration threshold, that time segment can be identified as a micro-arousal event.

[0062] Non-sleep events refer to segments of non-sleep states that occur during sleep, not belonging to any pre-defined sleep stage. Examples include prolonged wakefulness, getting out of bed, or other non-sleep states. Non-sleep events typically indicate significant sleep interruptions and have a substantial disruptive effect on sleep continuity. Non-sleep events can be identified using sleep staging data. When the stage label corresponding to a certain time period in the staging results does not belong to a pre-defined sleep stage category, the data segment corresponding to that time period can be identified as a non-sleep event.

[0063] A duration anomaly refers to a sleep stage that, while belonging to the normal sleep stage, has a significantly shorter continuous duration than expected, failing to meet the reference duration for that stage under normal sleep conditions, thus affecting the integrity of the sleep structure. This type of event can be identified by segmenting sleep stage data. After determining the actual duration of a sleep stage, this duration is compared to the corresponding reference duration. When the actual duration is less than the reference duration, the data segment corresponding to that sleep stage can be identified as a duration anomaly event.

[0064] Using the above method, different situations in sleep monitoring data that adversely affect sleep continuity can be uniformly identified as continuous abnormal events, laying the foundation for subsequent calculations of the impact of various abnormal events on the overall sleep process.

[0065] 103. Based on the time attributes and characteristic attributes of each continuous abnormal event, calculate the event impact time of each continuous abnormal event.

[0066] In this embodiment of the invention, the event impact time is used to represent the effective duration of a continuous abnormal event that disrupts sleep continuity. This duration is related not only to the duration of the abnormal event itself, but also to factors such as the timing of the abnormal event, the sleep stage in which it occurs, and the intensity of the event. Therefore, when calculating the event impact time, both time attributes and characteristic attributes need to be considered simultaneously.

[0067] The temporal attribute is mainly used to reflect the temporal location characteristics of abnormal events throughout the night's sleep process, such as the accumulated sleep duration before the abnormal event occurred and the duration of the abnormal event itself. Generally, the degree of impact on sleep continuity of abnormal events that occur in the later part of the night or after a long period of continuous sleep may differ from that of abnormal events that occur in the early stages of sleep. Therefore, it is necessary to include the temporal location of abnormal events during sleep in the calculation.

[0068] Feature attributes are primarily used to characterize the type and severity of abnormal events. Different types of continuous abnormal events correspond to different feature attributes. For example, micro-arousal events can be comprehensively evaluated by combining the sleep stage at the time of the micro-arousal, the degree of micro-arousal, and the weight corresponding to that sleep stage; non-sleep events can be evaluated by combining the position in the sleep cycle before the event occurs; and abnormal events of continuous duration of sleep stages can be evaluated by combining the actual duration of the stage, the reference duration, and sleep depth-related features.

[0069] In the specific calculation process of computer equipment, based on different types of continuous abnormal events, a preset impact time calculation rule can be used to combine the aforementioned time attributes and characteristic attributes to obtain the event impact time corresponding to the abnormal event. This event impact time is used to reflect the range of interference caused by the abnormal event to the subsequent sleep process, rather than being simply equivalent to the duration of the abnormal event itself.

[0070] For example, when a short but intense micro-arousal occurs during a deep sleep stage, its impact time may be significantly longer than the duration of the micro-arousal itself; while a similar micro-arousal occurring during a light sleep stage may have a relatively shorter impact time. In this way, the impact of different consecutive abnormal events on sleep continuity can be quantified on a uniform time scale, providing a basis for subsequent sleep continuity assessments.

[0071] 104. Based on the event impact time of each continuous abnormal event, determine the sleep continuity assessment result of the user to be evaluated.

[0072] In this embodiment of the invention, firstly, based on the event impact time of each continuous abnormal event, the time spent in a continuous abnormal state throughout the entire night's sleep is statistically analyzed to obtain the total duration of continuous abnormality. The time period of continuous anomalies can be obtained by marking and superimposing the event impact times of each continuous anomaly event.

[0073] Based on this, the total effective sleep duration used for assessment was further determined. The total effective sleep duration refers to the time interval between the start and end times of sleep. The start time is defined as the moment when a user first enters a sleep state and the duration of that sleep state exceeds a preset threshold TH7, such as 30 minutes, to avoid short naps being mistaken for formal sleep. The end time is defined as the end time of the last sleep segment that is not a nap. A nap refers to being in a sleep state but where the time interval between this sleep state and the most recent sleep state in the main sleep segment exceeds a preset threshold TH8, such as 3 hours.

[0074] The main sleep segment represents the data segment with the longest effective sleep duration throughout the night. When the duration of a non-sleep stage in the monitoring data exceeds the preset threshold TH9, such as 2 hours, it can be considered that the effective sleep state has been interrupted, and the non-sleep stage is marked as an invalid sleep state. The remaining continuous sleep data segments are included in the statistics as effective sleep segments.

[0075] Determining the total duration of continuous anomalies and total effective sleep duration After that, the sleep continuity index can be calculated. The calculation method is as follows:

[0076]

[0077] The sleep continuity index reflects the degree to which the continuity of sleep is disrupted throughout the night. The smaller the proportion of the total duration of continuity abnormalities in the total effective sleep time, the closer the calculated sleep continuity index is to 1, indicating a more stable and continuous sleep process. Conversely, when continuity abnormalities are frequent or have a long duration, the sleep continuity index decreases accordingly, reflecting poor sleep continuity.

[0078] In this way, continuous abnormal events of different types and locations can be uniformly mapped by computer equipment into a comprehensive impact on the continuity of sleep throughout the night, thereby achieving a quantitative assessment of sleep continuity.

[0079] In this embodiment of the invention, sleep monitoring data of the user to be evaluated is acquired; based on the sleep monitoring data, at least one continuous abnormal event is identified, wherein the continuous abnormal event is a micro-awakening event, a non-sleep event, or a duration abnormal event indicating a continuous duration abnormality in each sleep stage in the sleep monitoring data; based on the time attribute and feature attribute of each continuous abnormal event, the event impact time of each continuous abnormal event is calculated; based on the event impact time of each continuous abnormal event, the sleep continuity assessment result of the user to be evaluated is determined. By executing the steps of identifying different types of continuous abnormal events from sleep monitoring data and calculating the corresponding event impact time based on the time attribute and feature attribute of each continuous abnormal event, the computer device quantifies and integrates the impact of different abnormal situations on overall sleep in a time dimension, avoiding the shortcomings of rough assessments based solely on the number of events or duration. This allows the sleep continuity assessment result to more accurately reflect the actual degree of sleep continuity impairment, improving the objectivity and rationality of the sleep continuity assessment.

[0080] It is understood that in the specific embodiments of this application, data such as sleep monitoring data, user information, and historical sleep monitoring data are involved. When the embodiments of this application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0081] Optionally, the sleep monitoring data includes sleep event data. In the step of identifying at least one continuous abnormal event based on the sleep monitoring data, the occurrence time and end time of micro-awakening can also be identified in the sleep event data. When the duration from the occurrence time to the end time is less than a preset micro-awakening duration, the sleep segment from the occurrence time to the end time is identified as a micro-awakening event.

[0082] In this embodiment of the invention, the sleep event data includes at least information related to micro-arousals, which is used for refined identification and analysis of sleep continuity abnormalities.

[0083] Micro-awakening events are used to characterize brief awakening states that occur during sleep. A micro-awakening is defined as a segment of wakefulness that occurs during sleep, with a duration less than a preset threshold (e.g., 2 minutes), and both the preceding and following periods are within the sleep stage, thus eliminating interference from prolonged wakefulness or getting out of bed. Sleep event data can directly record the occurrence time, end time, and corresponding duration of micro-awakenings. When the duration between the occurrence and end time is less than a preset micro-awakening duration threshold, the corresponding sleep segment can be identified as a micro-awakening event.

[0084] In a further embodiment, the sleep event data may also include micro-arousal level information to characterize the intensity of micro-arousals. The micro-arousal level can be calculated based on heart rate variability characteristics, specifically using the ratio of low-frequency energy to high-frequency energy (LF / HF) as a characterization index. The sensor signal used to calculate this index can be a cardiac impulse signal or a photoplethysmography (PPG) signal. First, the computer device can filter the acquired sensor signal, for example, using a bandpass filter, to suppress noise and highlight cardiac-related components; then, specific feature points are extracted from the filtered signal, and these feature points can be selected as signal peaks.

[0085] After obtaining the feature points, the heartbeat interval sequence is calculated based on the time intervals between adjacent feature points, and anomaly removal and interpolation are performed on the heartbeat intervals. Anomaly removal can be performed using a sliding window method, calculating the mean heartbeat interval within a window time TH1, and removing abnormal heartbeat intervals that deviate from this mean by 20%. The window time TH1 can be 30 seconds. After anomaly removal, the remaining heartbeat intervals are interpolated, for example, using linear interpolation to convert the heartbeat interval sequence into an equal time interval sequence, such as one sampling point per second.

[0086] Subsequently, using the current time point as the endpoint, a spectral analysis was performed on a heart rate interval sequence with a forward time threshold TH2, which can be 5 minutes. Fourier transform was performed on this heart rate interval sequence to obtain the frequency domain energy distribution. The sum of spectral amplitudes within the frequency range of 0.04Hz to 0.15Hz was calculated as the low-frequency energy (LF); simultaneously, the sum of spectral amplitudes within the frequency range of 0.15Hz to 0.4Hz was calculated as the high-frequency energy (HF). By calculating the ratio of low-frequency energy to high-frequency energy (LF / HF), the heart rate variability characteristic used to characterize the state of autonomic nervous activity was obtained.

[0087] Based on this, the average LF / HF ratio during the corresponding sleep stage duration before the micro-awakening can be calculated, denoted as LF / HF_ref, and the mean LF / HF ratio during the micro-awakening time segment can be calculated, denoted as LF / HF_mean. The ratio of LF / HF_mean to LF / HF_ref is used as the degree of micro-awakening to reflect the intensity of physiological disturbance of the micro-awakening event relative to the normal sleep state.

[0088] After identifying micro-awakening events, further processing can be performed during the continuity anomaly marking process. Based on sleep event data, the occurrence time of micro-awakening events is located, and sleep continuity anomalies are marked at that time. Simultaneously, by combining sleep stage data, information such as the sleep stage preceding the micro-awakening, the cumulative sleep duration before the micro-awakening, the sleep cycle number of the micro-awakening, the duration of the micro-awakening, and the degree of micro-awakening are extracted to construct a micro-awakening feature array. The sleep duration before the micro-awakening refers to the sum of the time the user is in a sleep stage from the moment they first enter a sleep state to the moment the current micro-awakening begins.

[0089] To ensure the correspondence between micro-awakening events and specific sleep stages, the entire night's sleep data can be segmented. Starting from the beginning and ending point of each sleep stage, corresponding sleep stage data segments are formed, and feature information such as stage labels, sleep cycle numbers, and durations are extracted for each segment. Based on this, computer equipment performs anomaly detection on the sleep stage data segments, distinguishing between normal and abnormal sleep stages, thus providing a reliable data foundation for calculating the duration of subsequent continuous anomalies.

[0090] Through the above methods, micro-awakening events can not only be accurately identified, but also carry multi-dimensional feature information reflecting their location, duration, and intensity of physiological disturbance, providing a more refined and reliable basis for sleep continuity assessment.

[0091] Optionally, the sleep monitoring data includes sleep stage data. In the step of identifying at least one continuous abnormal event based on the sleep monitoring data, stage labels for each sleep stage can also be extracted from the sleep stage data. If the stage label is not a preset sleep event label, the corresponding sleep stage is a non-sleep event.

[0092] In this embodiment of the invention, sleep monitoring data may further include sleep staging data, used to provide a staged description of the entire night's sleep process. Sleep staging data identifies the user's sleep state within different time segments, and its stage labels may include wakefulness, various stages of non-rapid eye movement (NREM) sleep, and rapid eye movement (REM) sleep. By parsing the sleep staging data, the entire night's sleep process can be divided into several continuous sleep stage segments, each segment corresponding to a specific stage label and its start and end times.

[0093] In determining continuous abnormal events, stage labels for each sleep stage can be extracted based on sleep staging data, and these stage labels can be compared with preset sleep event labels. The preset sleep event labels represent the stage types considered part of a normal sleep process, such as non-rapid eye movement (NREM) sleep and rapid eye movement (REM) sleep. When the stage label corresponding to a certain stage segment does not belong to the preset sleep event labels, that stage segment can be considered not to be in a valid sleep state, thus identifying that stage segment as a non-sleep event.

[0094] Non-sleep events are used to characterize moments of wakefulness, getting out of bed, or other time segments that do not conform to the normal sleep stage definition during sleep. This stage-label-based determination method allows for the direct identification of non-sleep periods during sleep using sleep stage results without relying on additional behavioral sensors or manual annotation. For example, if a period of time during the entire night's sleep is staged as a wakeful state or cannot be categorized into any preset sleep stage label, that period can be automatically identified as a non-sleep event.

[0095] In further processing, the phase fragments corresponding to non-sleep events can be temporally correlated with the sleep stages before and after their occurrence to analyze the degree of impact of the non-sleep event on sleep continuity. By clearly distinguishing non-sleep events from the overall sleep stage sequence, computer equipment can avoid incorrectly including awake or interrupted sleep time in the effective sleep duration, thus providing a more accurate basis for calculating the impact time of subsequent continuity anomalies and determining the results of sleep continuity assessment.

[0096] Optionally, in the step of identifying at least one continuous abnormal event based on sleep monitoring data, if the stage label is a preset sleep event label, then the stage duration of the corresponding sleep stage can be determined; based on the stage duration, a continuous duration deviation judgment can be performed in a preset reference duration dataset to obtain a judgment result; when the judgment result is that the stage duration is less than the corresponding reference duration in the preset reference duration dataset, the corresponding sleep stage is a duration abnormal event.

[0097] In this embodiment of the invention, the duration of sleep stages belonging to the normal sleep type can be further analyzed to identify cases of abnormal continuous duration of sleep stages. This analysis is based on sleep stage data and is applicable to sleep stage segments whose stage labels belong to preset sleep event labels.

[0098] Specifically, when the stage tag corresponding to a certain sleep stage segment is determined to be a preset sleep event tag, the stage duration of that sleep stage can be determined first. The stage duration is obtained from the time interval between the start and end times of the sleep stage, and is used to reflect the actual continuous duration of the stage within the current sleep cycle.

[0099] After obtaining the stage duration, it can be compared with a preset reference duration dataset to determine if there is a significant deviation in the duration of the sleep stage. The reference duration dataset stores the reference duration of different sleep stages under normal sleep conditions, and this reference duration can be associated with sleep stage type, sleep cycle position, and individual user characteristics. By comparing the actual stage duration with the reference duration, the result of the duration deviation judgment can be obtained.

[0100] When the assessment results indicate that the actual duration of a certain sleep stage is less than the corresponding reference duration in the reference duration dataset, it can be considered that the sleep stage has insufficient continuous duration during the current sleep process, thus identifying the sleep stage as a duration aberration event. This type of event is used to characterize situations where, although the person is still in a sleep state, the sleep stage is maintained for too short a time and fails to meet the continuity requirements of normal sleep structure.

[0101] Using the above method, abnormalities in the continuous duration of sleep stages can be identified directly using sleep staging results and stage duration information without introducing additional sleep event annotations. These abnormal events can reflect frequent interruptions in sleep structure or insufficient sleep depth, providing a basis for subsequent calculations of the impact time of continuity abnormalities, thus making the sleep continuity assessment results more comprehensive and consistent with actual sleep quality characteristics.

[0102] Optionally, before the step of judging the continuous duration deviation based on the stage duration in the preset reference duration dataset and obtaining the judgment result, user information of the user to be evaluated and historical sleep monitoring data that meet the preset conditions can be obtained; based on the historical sleep monitoring data, the reference duration calculation coefficient of each sleep stage is calculated; based on the reference duration calculation coefficient of each sleep stage and the user information, the reference duration of each sleep stage is calculated; and a preset reference duration dataset is constructed based on the reference duration of each sleep stage.

[0103] In this embodiment of the invention, before judging the duration deviation of the sleep stage, a reference duration dataset for the judgment can be pre-constructed. This reference duration dataset is used to characterize the reasonable duration range of each sleep stage in different sleep cycles for different users under normal sleep continuity conditions.

[0104] Specifically, the process begins by acquiring user information and historical sleep monitoring data that meets preset criteria. User information includes at least age, gender, and body mass index (BMI), where BMI is calculated by dividing weight by the square of height. Historical sleep monitoring data consists of sleep data that has been staged and assessed using a sleep continuity scale after the sleep period, with the assessment result indicating good sleep continuity. This scale can be a preset standard scale or a custom scale, used to select representative, high-quality sleep samples.

[0105] After obtaining the aforementioned historical sleep monitoring data, the historical sleep data can be sliced ​​according to sleep cycles. The sleep cycle slicing method involves using the end time of each REM phase as the dividing boundary, dividing the entire night's sleep data into several sleep cycle segments, and removing non-sleep stages from each sleep cycle segment, thus obtaining sleep cycle data containing only effective sleep stages. Based on this processing result, the duration of different sleep stages within each sleep cycle can be calculated, and the average duration of each sleep stage in the corresponding sleep cycle can be further calculated.

[0106] Based on this, a pre-defined reference duration calculation model can be used to fit different sleep stages using the statistically obtained average duration, thereby obtaining the reference duration calculation coefficient for each sleep stage. Specifically, the reference duration for different sleep stages can be described using different mathematical forms.

[0107] For non-rapid eye movement (NREM) sleep stage 3 (N3), the reference duration is denoted as:

[0108]

[0109] Where i is the sleep cycle number. This is the coefficient for calculating the reference duration of N3 phase, where Age is age, Gender is sex, and BMI is body mass index, calculated by dividing weight by the square of height. It is a natural constant;

[0110] For non-rapid eye movement (NREM) sleep stage 2 (N2), the reference duration is denoted as:

[0111]

[0112] in The coefficient is used to calculate the reference duration length for period N2;

[0113] For non-rapid eye movement (NREM) sleep stage 1 (N1), the reference duration is denoted as:

[0114]

[0115] in The coefficient is used to calculate the reference duration length for period N1.

[0116] For REM sleep, the reference duration is denoted as:

[0117]

[0118] in The coefficient is used to calculate the reference duration of the REM period, and ln() is the natural logarithm.

[0119] The reference duration coefficients for each sleep stage can be obtained through statistical analysis and parameter fitting of historical sleep monitoring data that meet preset conditions. During the statistical process, the average duration of each sleep stage within each sleep cycle can be used as the fitting target. Based on the functional relationship in the corresponding reference duration model, regression fitting is performed on the calculated coefficients to obtain reference duration coefficients that reflect the continuity characteristics of normal sleep.

[0120] After obtaining the reference duration calculation coefficients for each sleep stage, the computer device can combine the user information of the user to be evaluated, substituting the user's age, gender, and BMI into the corresponding reference duration calculation formula, to calculate the reference duration for each sleep stage under different sleep cycles. Finally, the reference durations for each sleep stage under different sleep cycles can be uniformly compiled to construct a preset reference duration dataset, which is used for subsequent judgment of continuous duration deviations from actual sleep stages, thereby identifying abnormal events in continuous sleep stage duration.

[0121] Optionally, in the step of calculating the event impact time of each continuous abnormal event based on its temporal and characteristic attributes, for micro-awakening events, the temporal attributes of micro-awakening events include the sleep duration before the micro-awakening event and the duration of the micro-awakening event. The characteristic attributes include the sleep cycle number where the micro-awakening event occurs, the degree of micro-awakening in the micro-awakening event, and the weight of the sleep stage before the micro-awakening event occurs. The weight is set by the statistical probability of the micro-awakening event in each sleep stage. Alternatively, the event impact time of micro-awakening events can be calculated based on the sleep duration before the micro-awakening event occurs, the duration of the micro-awakening event, the sleep cycle number where the micro-awakening event occurs, the degree of micro-awakening in the micro-awakening event, and the weight of the sleep stage before the micro-awakening event occurs.

[0122] In this embodiment of the invention, for continuous abnormal events caused by micro-awakening, the influence range of micro-awakening on the overall sleep continuity can be quantitatively characterized by combining time attributes and feature attributes, and the time interval of the continuous abnormality can be determined accordingly.

[0123] Combination Figure 2 One of the diagrams illustrating the sleep staging results shows that the entire night's sleep is divided into multiple sleep cycles, with different sleep stages alternating on the timeline. These different sleep stages reflect the differences in the user's physiological state during sleep. Specifically, these include: non-rapid eye movement (NREM) sleep stage 3 (N3), NREM sleep stage 2 (N2), NREM sleep stage 1 (N1), rapid eye movement (REM) sleep stage, wakefulness stage (AWAKE), and absent stage (absent from bed or signal loss).

[0124] Among them, N3 is the deep sleep stage, with the highest sleep depth and the lowest level of physiological activity, and usually has a strong restorative effect; N2 is the steady sleep stage, which accounts for the largest proportion of the sleep throughout the night, and is between light sleep and deep sleep; N1 is the sleep onset or light sleep stage, which often occurs during the transition from wakefulness to steady sleep, and has a lower sleep depth; REM is the rapid eye movement sleep stage, in which brain electrical activity is close to the wakefulness state, usually related to dreaming activity, and appears periodically in each sleep cycle.

[0125] AWAKE indicates that the user is awake, meaning that there is a brief or continuous awakening during sleep; ABSENT indicates the period of time when the user gets out of bed or when the sleep monitoring signal is interrupted.

[0126] Micro-awakenings typically manifest as short-lived arousal fragments that occur within a sleep stage or near stage transitions. Although these fragments are brief, they can disrupt sleep structure within a certain timeframe before and after them, thus requiring extended labeling by influencing their timing.

[0127] When a series of abnormal events are micro-awakening events, the first step is to obtain the corresponding time attributes of the micro-awakening events, including the cumulative sleep duration before the occurrence of the micro-awakening event. And the duration of the micro-awakening event itself. The cumulative sleep duration refers to the total time spent in a sleep state from the moment one enters effective sleep to the moment that micro-awakening begins. Combined with... Figure 2 The mid-time axis can be understood as reflecting the relative position of micro-awakenings throughout the night's sleep; the closer to the later part of sleep, the smaller their weight in affecting the overall continuity.

[0128] Simultaneously, feature attributes related to micro-arousals can be extracted, including the sleep cycle number C of the micro-arousal event, the degree of micro-arousal ω, and the weight S corresponding to the sleep stage before the micro-arousal. The sleep cycle number distinguishes which complete sleep cycle the micro-arousal occurs within; the cycle structure separated by dashed lines in the figure visually illustrates this information. The degree of micro-arousal ω characterizes the strength of the physiological disturbance caused by the micro-arousal, which can be obtained by the ratio of LF / HF_mean to LF / HF_ref, reflecting the magnitude of change in autonomic nervous activity relative to the stable sleep stage during the micro-arousal. The weight S corresponding to the sleep stage characterizes the difference in the impact of micro-arousals at different sleep depths. This weight can be set based on the statistical probability of micro-arousals in each sleep stage under complete sleep conditions. The probability of micro-arousals is lower in deep sleep stages, so the corresponding weight is larger; for example, 2 for N3, 1.5 for N2, 1 for REM sleep, and 0.5 for N1.

[0129] Understandably, the weights of sleep stages corresponding to micro-arousals are not determined by the order of sleep stages in the sleep cycle. The reason for considering the statistical probability of micro-arousals occurring in each sleep stage under conditions of complete and stable sleep is that, generally, the lower the probability of micro-arousals in a sleep stage during complete sleep, the higher the stability that stage should maintain. Once a micro-arousal occurs, the greater the disturbance and disruption to sleep continuity, and therefore the higher the corresponding weight value. Conversely, in stages with a higher probability of micro-arousals and less stable sleep, the disruption to continuity is relatively smaller, and the corresponding weight value is lower. Therefore, we can set, for example, a weight of 2 for N3, 1.5 for N2, 1 for REM, and 0.5 for N1, to accurately reflect the differences in the impact of micro-arousals on sleep continuity at different sleep depths.

[0130] After considering the aforementioned time and characteristic attributes, the duration of influence of micro-awakening events can be calculated according to the following rules. :

[0131]

[0132] in, , and The coefficients for calculating the duration of micro-arousals are used to adjust the relative contributions of different factors in the calculation of the duration of influence. The exponential term describes the gradual decay of the impact of a single micro-arousal on the overall sleep continuity as the cumulative sleep duration and sleep cycle number increase; the weight S and the micro-arousal intensity coefficient ω are used to amplify the impact of micro-arousal events that occur during deep sleep or when physiological disturbances are strong. It directly reflects the linear contribution of the duration of micro-awakening to the range of influence.

[0133] The above calculated coefficients , and This can be obtained through statistical fitting of a specific dataset. This specific dataset consists of data that has been labeled with sleep stages, contains no awakening or out-of-bed segments during sleep, and has been assessed using a sleep continuity scale after the sleep period. In the statistical process, the total duration of micro-awakenings throughout the night can be estimated using an empirical model based on the scale assessment results. Then, the feature array corresponding to each micro-awakening segment is extracted, including the sleep duration before the micro-awakening, the sleep cycle number where the micro-awakening occurs, the weight of the sleep stage before the micro-awakening, the micro-awakening intensity coefficient, and the duration of the micro-awakening, and then... To determine the target value, the parameters of the above influence time model are fitted. , and The value of .

[0134] During the period of micro-awakening influence After calculation, the starting point of micro-awakening can be taken as the starting position, and the period after that point can be used as the starting position. The time segments were uniformly marked as continuous anomalous intervals. Combined with... Figure 2 The continuity status bar in the diagram can be understood as follows: the instantaneous location of a micro-awakening corresponds to the marked starting point on the status bar, while the extended period afterward reflects the continuous impact of micro-awakening on the subsequent sleep structure. This period of impact is uniformly included in the continuity abnormality duration in the continuity assessment, thus more realistically reflecting the degree of disruption of the continuity of sleep throughout the night by micro-awakening.

[0135] Optionally, in the step of calculating the event impact time of each continuous abnormal event based on the time attribute and feature attribute of each continuous abnormal event, for non-sleep events, the time attribute includes the sleep duration before the occurrence of the non-sleep event, and the feature attribute includes the sleep cycle number before the occurrence of the non-sleep event; the event impact time of non-sleep events can also be calculated based on the sleep duration before the occurrence of the non-sleep event and the sleep cycle number before the occurrence of the non-sleep event.

[0136] In this embodiment of the invention, when a continuous abnormal event is caused by a non-sleep stage, it is necessary to quantify the range of disturbance of the non-sleep stage on the preceding and following sleep structures, thereby obtaining the event impact time of the non-sleep event, and completing the continuous abnormality marking on the timeline. The non-sleep stage can be understood as a segment of the sleep staging results that does not belong to the preset sleep event label, such as awakening, getting out of bed, etc., which are not counted as effective sleep stages; once such a segment occurs, it is often not only an interruption itself, but also has a continuous impact on the sleep stability before and after it. Therefore, the event impact time is used to extend the characterization of this impact.

[0137] The temporal attribute of non-sleep events is the cumulative sleep duration prior to the occurrence of the non-sleep event. The cumulative sleep duration is used to reflect the relative position of the non-sleep event within the entire night's sleep; the characteristic attribute uses the sleep cycle number before the non-sleep event occurred. The sleep cycle number is used to indicate which sleep cycle the non-sleep event occurred in. Combined with... Figure 3 The second diagram of the sleep staging results shows that the segment corresponding to the non-sleep stage is located in the middle of the sleep staging sequence. The start and end positions of the non-sleep stage are marked with "1" and "2" below the continuity status bar. "1" corresponds to the start time of the non-sleep stage, and "2" corresponds to the end time of the non-sleep stage. On this basis, the markers affecting time do not only cover the non-sleep segment between "1" and "2", but extend to both sides from "1" and "2" as the center, covering a wider time range to reflect the continuous disturbance caused by non-sleep.

[0138] The duration of the event's impact during non-sleep phases is denoted as... The calculation rules are as follows:

[0139]

[0140] in, The coefficient for calculating the impact time during the non-sleep phase; This refers to the cumulative sleep duration before the onset of this non-sleep phase. This represents the sleep cycle number preceding the non-sleep stage; the abnormal event type is non-sleep stage. This model reflects two intuitions: the closer a non-sleep event is to the latter part of the night's sleep, or the more likely it is to occur at different points in the sleep cycle, the greater the degree of disturbance to the overall continuity may be. Therefore, by… and Joint regulation , and by Provides a reference bias.

[0141] This can be obtained by statistically fitting a specific dataset 2. Specific dataset 2 consists of historical sleep data with pre-defined sleep staging results, meaning the sleep data has been staged and meets relatively stable sleep structure constraints: the total number of micro-awakenings is less than a preset threshold TH5 (e.g., TH5 = 40 times); the number of sleep stage transitions within each sleep cycle is less than a preset threshold TH6 (e.g., TH6 = 8 times); the sleep duration reaches a specified duration (e.g., 7 hours); and a sleep continuity scale assessment is performed after sleep. During the statistical process, an empirical model is used to estimate the total duration of sleep affected by non-sleep factors throughout the night based on the scale assessment results. The system extracts feature arrays corresponding to each non-sleep segment during the entire night's sleep process. The feature arrays include at least the sleep duration before the non-sleep stage occurs. Sleep cycle number before the non-sleep stage Subsequently, based on the above... The mapping relationship, with Parameters are fitted to the target value to determine The value of .

[0142] After completion After calculation, the time markers for the impact of the non-sleep phase are implemented by symmetrically expanding around the start and end points of the non-sleep segment. Specifically, this involves starting before the beginning of the non-sleep phase... From the point in time until after the end of the non-sleep phase. Up to the specified point in time, this time interval is uniformly marked as a continuous anomaly. Combined with... Figure 3 The continuity status bar shown extends to the left from the starting point marked "1". And the end point of the "2" mark extends to the right. The expanded interval is the interval of continuous abnormal effects caused by non-sleep events; Figure 3 The data only shows continuous abnormality markers caused by non-sleep stages, so that the affected intervals are aligned one-to-one with the sleep stage sequence on the time axis and can be used later to accumulate the total duration of continuous abnormalities.

[0143] Optionally, in the step of calculating the event impact time of each continuous abnormal event based on the time attribute and feature attribute of each continuous abnormal event, for duration abnormal events, the time attribute includes the actual duration of the sleep stage corresponding to the duration abnormal event and the reference duration of the sleep stage, and the feature attribute includes the sleep cycle number of the sleep stage and the weight of the sleep stage, with the weight set according to the sleep depth of the user to be evaluated; the event impact time of the duration abnormal event can also be calculated based on the actual duration of the sleep stage corresponding to the duration abnormal event, the reference duration of the sleep stage, the sleep cycle number of the sleep stage, and the weight of the sleep stage.

[0144] In this embodiment of the invention, when a continuity anomaly is caused by an excessively short consecutive sleep stage, the excessively short sleep stage is regarded as a "duration anomaly event," and its disturbance range on the continuity of subsequent sleep is further quantified to obtain the event impact time of the duration anomaly event. Simultaneously, continuity anomaly marking is completed on the timeline. The determination of duration anomalies is based on stage slices of sleep stage data and reference duration thresholds: first, the actual duration of each sleep stage is obtained from the stage slices, and then the reference duration threshold at the same sleep cycle position is obtained from the reference duration dataset. When the actual duration is significantly less than the reference duration threshold, it is considered that there is an abnormal risk of excessively short continuous duration. The key point here is not to end after "identifying" the abnormality, but to "expand" the impact of the abnormality for subsequent cumulative total duration of continuous abnormalities.

[0145] The duration of an abnormal event is recorded as the event impact time. The calculation rules are as follows:

[0146]

[0147] in, This refers to the actual duration of the sleep stage; This is the reference duration threshold for this sleep stage, and its acquisition process corresponds to the reference duration acquisition step in the stage segment anomaly judgment. A coefficient used to calculate the impact of insufficient continuous sleep duration on time; This refers to the sleep cycle number in which the sleep stage belongs; These are weighting coefficients corresponding to abnormalities in sleep stages, used to reflect the different sensitivities of sleep depth differences to continuous perturbations. The deeper the sleep, the larger the value. These values ​​can be set empirically, for example, 0.5 for N1, 1 for REM, 1 for N2, and 2 for N3. The meaning of this model can be understood from three dimensions: firstly, based on... The term expresses the "deviation between the actual and the reference value." The greater the deviation, the stronger the potential disturbance of the continuity anomaly; secondly, it is an exponential decay term. This is used to adjust the suppression or amplification of the influence range by different sleep cycle positions and relative deviation ratios, so that the same duration deviation can correspond to different influence times in different sleep cycles; thirdly... By further differentiating the impact intensity of different stages through sleep depth weighting, the impact on overall continuity is more significant when the continuity of deep sleep is disrupted.

[0148] This can be obtained by statistically fitting a specific dataset 3. Specific dataset 3 meets the following criteria: sleep data includes sleep stage results and is fully labeled; there are no awake or out-of-bed time segments during sleep; the total number of micro-awakenings is less than a preset threshold TH5; sleep duration reaches a specified duration, such as 7 hours; and a sleep continuity scale assessment was performed after sleep. During the statistical process, based on the scale assessment results, an empirical model is first used to estimate the total duration of impact caused by insufficient sleep stage continuity throughout the night. Then, feature arrays of data segments with excessively short durations are extracted segment by segment throughout the night's sleep. The feature arrays must include at least the duration of each sleep stage. Reference duration threshold for sleep stages The sleep stage and its corresponding sleep cycle number Weighting coefficients corresponding to sleep stage abnormalities .based on The mapping relationship between the above features, in order to Parameters are fitted to the target value to determine The value of is chosen so that the time distribution of the impact in the model output can be consistent with the overall duration of impact assessed by the scale.

[0149] After completion After calculation, the impact time stamp of the duration abnormal event is implemented by "extending backward from the end point of the abnormal phase": taking the end point of the continuous short sleep phase as the starting boundary, and extending from that end point to the point after that end point. Within a given timeframe, the continuity of the corresponding time segment is marked as a continuity anomaly. Combined with... Figure 4 The third diagram illustrating the sleep staging results clearly shows that the key locations of different short duration events are indicated by markers such as "1" and "2" below the continuity status bar. "1" corresponds to the end point of N3 (short duration N3), and "2" corresponds to the end point of N2 (short duration N2). The continuous interval following these markers represents the period after the sleep staging process. The extended range of continuous anomalies. Figure 4 The system only displays continuity anomalies caused by excessively short consecutive sleep stages, allowing the affected intervals of these anomalies to be directly added to the total duration of continuity anomalies for subsequent calculation of the sleep continuity index.

[0150] It should be noted that any of the sleep assessment methods provided in the embodiments of the present invention can be executed by a computer device using a computer program stored in a memory to analyze and calculate sleep monitoring data, thereby obtaining sleep continuity assessment results.

[0151] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0152] In one embodiment, a sleep assessment device is provided, which corresponds one-to-one with the sleep assessment methods described in the above embodiments. For example... Figure 5 As shown, the sleep assessment device includes a first acquisition module 501, a first determination module 502, a first calculation module 503, and a second determination module 504. Detailed descriptions of each functional module are as follows:

[0153] The first acquisition module 501 is used to acquire sleep monitoring data of the user to be evaluated;

[0154] The first determining module 502 is used to determine at least one continuous abnormal event based on the sleep monitoring data. The continuous abnormal event is a micro-awakening event, a non-sleep event, or a duration abnormal event in which the duration of each sleep stage in the sleep monitoring data is abnormal.

[0155] The first calculation module 503 is used to calculate the event impact time of each of the continuous abnormal events based on the time attribute and feature attribute of each of the continuous abnormal events.

[0156] The second determining module 504 is used to determine the sleep continuity assessment result of the user to be evaluated based on the event impact time of each of the continuous abnormal events.

[0157] Optionally, the sleep monitoring data includes sleep event data, and the first determining module 502 is further configured to:

[0158] The occurrence and end times of micro-awakening were determined from the sleep event data.

[0159] When the duration from the occurrence time point to the end time point is less than a preset micro-awakening duration, the sleep segment from the occurrence time point to the end time point is determined as the micro-awakening event.

[0160] Optionally, the sleep monitoring data includes sleep stage data, and the first determining module 502 is further configured to:

[0161] Stage labels for each sleep stage are extracted from the sleep stage data;

[0162] If the stage label is not a preset sleep event label, then the corresponding sleep stage is the non-sleep event.

[0163] Optionally, the first determining module 502 is further configured to:

[0164] If the stage label is a preset sleep event label, then the stage duration corresponding to the sleep stage is determined;

[0165] Based on the duration of the aforementioned stage, a continuous duration deviation is judged in a preset reference duration dataset to obtain a judgment result;

[0166] When the determination result is that the duration of the stage is less than the corresponding reference duration in the preset reference duration dataset, the corresponding sleep stage is a duration abnormal event.

[0167] Optionally, the device further includes:

[0168] The second acquisition module is used to acquire user information of the user to be evaluated and historical sleep monitoring data that meet preset conditions;

[0169] The first statistical module is used to calculate the reference duration coefficient for each sleep stage based on the historical sleep monitoring data.

[0170] The second calculation module is used to calculate the reference duration of each sleep stage based on the reference duration calculation coefficient of each sleep stage and the user information.

[0171] The first construction module is used to construct the preset reference duration dataset based on the reference duration of each sleep stage.

[0172] Optionally, the first computing module 503 is further used for

[0173] For the micro-awakening event, the time attribute of the micro-awakening event includes the sleep duration before the occurrence of the micro-awakening event and the duration of the micro-awakening event. The feature attribute includes the sleep cycle number where the micro-awakening event occurs, the degree of micro-awakening of the micro-awakening event, and the weight corresponding to the sleep stage before the occurrence of the micro-awakening event. The weight is set by the statistical probability of the micro-awakening event in each sleep stage.

[0174] The event impact time of the micro-awakening event is calculated based on the sleep duration before the micro-awakening event, the duration of the micro-awakening event, the sleep cycle number in which the micro-awakening event occurs, the degree of micro-awakening of the micro-awakening event, and the weight of the sleep stage before the micro-awakening event.

[0175] Optionally, the first computing module 503 is further used for

[0176] For the non-sleep event, the time attribute includes the sleep duration before the non-sleep event occurred, and the feature attribute includes the sleep cycle number before the non-sleep event occurred.

[0177] The duration of sleep preceding the non-sleep event and the sleep cycle number preceding the non-sleep event are used to calculate the event impact time of the non-sleep event.

[0178] Optionally, the first computing module 503 is further used for

[0179] For the duration abnormal event, the time attribute includes the actual duration of the sleep stage corresponding to the duration abnormal event and the reference duration of the sleep stage. The feature attribute includes the sleep cycle number of the sleep stage and the weight corresponding to the sleep stage. The weight is set according to the sleep depth of the user to be evaluated.

[0180] The event impact time of the duration abnormal event is calculated based on the actual duration of the sleep stage corresponding to the duration abnormal event, the reference duration of the sleep stage, the sleep cycle number of the sleep stage, and the weight of the sleep stage.

[0181] For specific limitations regarding the sleep assessment device, please refer to the limitations on sleep assessment methods above, which will not be repeated here. Each module in the aforementioned sleep assessment device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0182] In one embodiment, a computer device is provided, which may be a terminal device, and its internal structure diagram may be as follows: Figure 6 As shown, the computer device includes a processor, memory, and a network interface connected via a system bus. The processor provides computing and control capabilities. The memory includes a readable storage medium storing computer-readable instructions. The network interface communicates with external terminals via a network connection. When executed by the processor, the computer-readable instructions implement a sleep evaluation method. The readable storage medium provided in this embodiment includes both non-volatile and volatile readable storage media.

[0183] In one embodiment of the application, a readable storage medium is provided, which stores computer-readable instructions. When the computer-readable instructions are executed by a processor, they implement the steps of the sleep assessment method described above.

[0184] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by instructing related hardware with computer-readable instructions. These computer-readable instructions can be stored in a non-volatile readable storage medium or a volatile readable storage medium. When executed, these computer-readable instructions can include the processes of the embodiments of the methods described above. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0185] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0186] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A sleep assessment method, characterized in that, The method includes: Obtain sleep monitoring data from the user to be evaluated; Based on the sleep monitoring data, at least one continuous abnormal event is identified. The continuous abnormal event is a micro-awakening event, a non-sleep event, or a duration abnormal event in which the duration of each sleep stage in the sleep monitoring data is abnormal. Based on the time attribute and characteristic attribute of each of the continuous abnormal events, the event impact time of each of the continuous abnormal events is calculated; Based on the event impact time of each of the continuous abnormal events, the sleep continuity assessment result of the user to be evaluated is determined; The sleep monitoring data includes sleep event data, and the determination of at least one continuous abnormal event based on the sleep monitoring data includes: The occurrence and end times of micro-awakening were determined from the sleep event data. When the duration from the occurrence time to the end time is less than a preset micro-awakening duration, the sleep segment from the occurrence time to the end time is determined as the micro-awakening event; The sleep monitoring data includes sleep stage data, and the determination of at least one continuous abnormal event based on the sleep monitoring data includes: Stage labels for each sleep stage are extracted from the sleep stage data; If the stage label is not a preset sleep event label, then the corresponding sleep stage is the non-sleep event; The determination of at least one continuous abnormal event based on the sleep monitoring data further includes: If the stage label is a preset sleep event label, then the stage duration corresponding to the sleep stage is determined; Based on the duration of the aforementioned stage, a continuous duration deviation is judged in a preset reference duration dataset to obtain a judgment result; When the determination result is that the duration of the stage is less than the corresponding reference duration in the preset reference duration dataset, the corresponding sleep stage is a duration abnormal event. The step of calculating the event impact time of each of the continuous abnormal events based on the time attributes and feature attributes of each event includes: For the micro-awakening event, the time attribute of the micro-awakening event includes the sleep duration before the occurrence of the micro-awakening event and the duration of the micro-awakening event. The feature attribute includes the sleep cycle number where the micro-awakening event occurs, the degree of micro-awakening of the micro-awakening event, and the weight corresponding to the sleep stage before the occurrence of the micro-awakening event. The weight is set by the statistical probability of the micro-awakening event in each sleep stage. Based on the sleep duration before the occurrence of the micro-awakening event, the duration of the micro-awakening event, the sleep cycle number in which the micro-awakening event occurs, the degree of micro-awakening in the micro-awakening event, and the weight of the sleep stage before the occurrence of the micro-awakening event, the event impact time of the micro-awakening event is calculated. For the non-sleep event, the time attribute includes the sleep duration before the non-sleep event occurred, and the feature attribute includes the sleep cycle number before the non-sleep event occurred. Based on the sleep duration before the non-sleep event and the sleep cycle number before the non-sleep event, the event impact time of the non-sleep event is calculated. For the duration abnormal event, the time attribute includes the actual duration of the sleep stage corresponding to the duration abnormal event and the reference duration of the sleep stage. The feature attribute includes the sleep cycle number of the sleep stage and the weight corresponding to the sleep stage. The weight is set according to the sleep depth of the user to be evaluated. The event impact time of the duration abnormal event is calculated based on the actual duration of the sleep stage corresponding to the duration abnormal event, the reference duration of the sleep stage, the sleep cycle number of the sleep stage, and the weight of the sleep stage.

2. The sleep assessment method as described in claim 1, characterized in that, Before determining the duration deviation based on the stage duration in a preset reference duration dataset and obtaining the determination result, the method further includes: Obtain the user information of the user to be evaluated and historical sleep monitoring data that meet preset conditions; Based on the historical sleep monitoring data, the reference duration calculation coefficients for each sleep stage were statistically obtained. Based on the reference duration calculation coefficients for each sleep stage and the user information, the reference duration for each sleep stage is calculated. The preset reference duration dataset is constructed based on the reference duration of each sleep stage.

3. A sleep assessment device, characterized in that, The device includes: The first acquisition module is used to acquire sleep monitoring data of the user to be evaluated; The first determining module is used to determine at least one continuous abnormal event based on the sleep monitoring data. The continuous abnormal event is a micro-awakening event, a non-sleep event, or a duration abnormal event of continuous duration of each sleep stage in the sleep monitoring data. The first calculation module is used to calculate the event impact time of each of the continuous abnormal events based on the time attribute and feature attribute of each of the continuous abnormal events; The second determining module is used to determine the sleep continuity assessment result of the user to be assessed based on the event impact time of each of the continuous abnormal events; The sleep monitoring data includes sleep event data, and the first determining module is further configured to: The occurrence and end times of micro-awakening were determined from the sleep event data. When the duration from the occurrence time to the end time is less than a preset micro-awakening duration, the sleep segment from the occurrence time to the end time is determined as the micro-awakening event; The sleep monitoring data includes sleep stage data, and the first determining module is further configured to: Stage labels for each sleep stage are extracted from the sleep stage data; If the stage label is not a preset sleep event label, then the corresponding sleep stage is the non-sleep event; The first determining module is further configured to: If the stage label is a preset sleep event label, then the stage duration corresponding to the sleep stage is determined; Based on the duration of the aforementioned stage, a continuous duration deviation is judged in a preset reference duration dataset to obtain a judgment result; When the determination result is that the duration of the stage is less than the corresponding reference duration in the preset reference duration dataset, the corresponding sleep stage is a duration abnormal event. The first calculation module is further configured to: For the micro-awakening event, the time attribute of the micro-awakening event includes the sleep duration before the occurrence of the micro-awakening event and the duration of the micro-awakening event. The feature attribute includes the sleep cycle number where the micro-awakening event occurs, the degree of micro-awakening of the micro-awakening event, and the weight corresponding to the sleep stage before the occurrence of the micro-awakening event. The weight is set by the statistical probability of the micro-awakening event in each sleep stage. Based on the sleep duration before the occurrence of the micro-awakening event, the duration of the micro-awakening event, the sleep cycle number in which the micro-awakening event occurs, the degree of micro-awakening in the micro-awakening event, and the weight of the sleep stage before the occurrence of the micro-awakening event, the event impact time of the micro-awakening event is calculated. For the non-sleep event, the time attribute includes the sleep duration before the non-sleep event occurred, and the feature attribute includes the sleep cycle number before the non-sleep event occurred. Based on the sleep duration before the non-sleep event and the sleep cycle number before the non-sleep event, the event impact time of the non-sleep event is calculated. For the duration abnormal event, the time attribute includes the actual duration of the sleep stage corresponding to the duration abnormal event and the reference duration of the sleep stage. The feature attribute includes the sleep cycle number of the sleep stage and the weight corresponding to the sleep stage. The weight is set according to the sleep depth of the user to be evaluated. The event impact time of the duration abnormal event is calculated based on the actual duration of the sleep stage corresponding to the duration abnormal event, the reference duration of the sleep stage, the sleep cycle number of the sleep stage, and the weight of the sleep stage.

4. A computer device comprising a memory, a processor, and computer-readable instructions stored in the memory and running on the processor, characterized in that, When the processor executes the computer-readable instructions, it implements the sleep assessment method as described in any one of claims 1 to 2.

Images